Probabilistic Graphical Models (PGMs) are:

* Connectionist: RBMs are PGMs and neural networks ([source](http://image.diku.dk/igel/paper/AItRBM-proof.pdf))
* Bayesian: Bayes Networks are bayesian ([Wikipedia article](https://en.wikipedia.org/wiki/Bayesian_network))
* Symbolist: Markov Logic Networks ([source](http://link.springer.com/article/10.1007/s10994-006-5833-1))
* Analogizers and Evolutionaries: [According to Domingos](http://www.kdnuggets.com/2015/11/domingos-5-tribes-machine-learning-acm-webinar.html), they are also in Markov Logic Networks.

So the answer is that you can't simply categorize such a general technique as probabilistic graphical models in a single one of those categories.

See also: https://www.youtube.com/watch?v=E8rOVwKQ5-8